I was talking about Web Intelligence and Big Data

but I actually ended up unenrolling because I thought they would be pretty similar and I just dont have the time right now (mainly because I was in Mexico for a week and got fairly behind in Calculus) to do the "Big Data" course. So yah, should be doing Intro to Data Science still hopefully.

anyone take the MicroEconomics Principles for Univeristy of Illinois or Principles of Micro from UPENN with Coursera? Probably going to sign up for one of those since it has a workload that's manageable with my current schedule and starting soon

Also considering intro to interactive python from rice.

No, I signed up for it last year before I had any mathematical background and got stuck immediately. It is for this class and a few others that I'm wondering why LA isn't offered yet and I'm just going to dive in and do the OCW version soon.

The final for the UPenn Single Variable Calculus class is finally coming up. Just one more exam on Discrete Calculus before that. I'm not confident that I'll pass this course to my satisfaction if I can pass it at all. While building that poker bot was a lot of fun, it wasn't the wisest use of my time, but I was already comfortable with the reality that I'll have to take this class again.

Final tally:

5 exams
5 answer PDFs with 10+ pages of step-by-step explanations
1 final
56(!) homework assignments

If anyone wants to really really master Single Variable Calculus and it's applications to engineering and computer science, this class offers no excuses. Don't be put off by the hand-drawings: the course is absolutely top-notch. The effort put into the class is nothing less than monumental and I feel kind of guilty for not mastering all of the material, but there was legitimate time constraints and a busy schedule during the 3rd module that pushed me back a bit, plus there were many things that even they admitted was only for the truly brave. So, I'll take the final and hopefully get the certificate, but I'm sort of on the fence about actually getting a certificate because I do want to take the class again and I fear that gaining the cert will be a strong disincentive to not retaking the course.

Regardless, I'm very happy that this class is over with. Calculus is, IMO, pure torture and I have zero desire to explore multi-variable calculus.

Why is a calculus class a better use of time than a building a poker bot? It seems like the bot would cover more useful skills. I dropped out midway through the BerkleyAI course because something covered in the course solved a roadblock on one of my projects, and I figured the time would be better spent there. I never regretted it.

It's been 15 years, but I don't remember multi-variable calculus to be that hard at my college. Granted, It wasn't the most rigorous school, but the calculus class before multi-variable covered sequences and series and had a high attrition rate, and for good reason. I agree that calculus isn't much fun though and I wouldn't forced myself though a course without a definite application in mind.

I have zero education (HS dropout from top 5 poorest school district in the nation, fwiw), and this is about the only opportunity I have to prove my metal to the world and to myself, really. I'm actually kind of glad that I was able to teach myself enough Calc One from the OCW to understand what the hell is happening in this course (it's a Calc Two). The accomplishment is fine for itself, but I'd like some proof others can see, not what's in my own mind.

I'm going to keep those PDFs, download some of the videos, and copy/paste the homeworks so I can take my time to get it all right and destroy the class in the next iteration. I would consider doing Multi-Variable if the same lecturer does the course, but that won't happen any time soon. Rather concentrate on doing Linear Algebra and Discrete Math, anyways.

Why not MVC? Integrating is seriously disgusting to do and I hate it with an utter passion. Why would I want to do double and triple integration?

Whew, glad I copied that before the submit... I've lost posts here with the short timeout and unreliable re-signin.

It's interesting to see different people's take on education. I have a smart friend (probably 3+ SDs) who didn't finish high school, and he has always been self conscious about that.

One of my parents finished high school but didn't go to college, and the other didn't finish high school. One of my grandparents didn't even make it to high school; dropped out to pick cotton. In my family, being smart was valued, but formal education wasn't emphasized. It was assumed I would go to college, but no one was in a position to see the credentialling arms race that was really getting underway, and the growing value of a good school. (Then I had the poor timing to graduate into the '01 tech downturn, but that's another story.)

I said all that to say that, in my mind, it's more about what you can do rather than a credential, and in particular, what you can do that is valuable enough for someone to pay for. And working on a bot does more on these fronts than studying calculus.

I've pondered that stuff a lot over the past couple of years.

There was a real reason why I dropped out. It had nothing to do with my efforts and IQ, and I most certainly am not open about this to people in my real life, but there is still that ugly stigma attached to dropping out of HS, and that, more than anything, whether it describes me or not, eats at my conscience.

I can actually get past that a little bit, though. The ugly part is how judgmental people are towards people for not going to college, ie "If you don't care enough about your LIFE to spend a few years in college, then, ugh..." So many stupid opportunities were denied me because of this and it eats at me more than anything. My old boss seriously thought I had a masters in math or business or something like that, but lo and behold... While the idea that I give this impression feels good, it doesn't really offer me any real respect when people see nothing next to my name on a sheet of paper.

The stigma of being a dropout (and not attending college), is the assumption that I am a quitter, and that label hurts me more than anything, and this is why I won't accept failure. I don't want to accept myself as a quitter, and that is exactly what not gaining a certificate represents no matter how much amazing knowledge I gain or how I ultimately apply it or how visible / valuable anything else I do is or becomes. Passing with a 55% is also failure in my mind, so even the certificate isn't good enough for me.

In the summer time, I'll reach 2 years of learning programming, and I think year 3 will be focused on gaining at least 10 certificates. Whether this is a good strategy or not, I don't know and honestly don't care. If I had full confidence in getting a job, I'm sure I could have landed one by now. If people who don't know anything beyond installing scripts and are landing jobs after "learning to program" in less than 12 months, I'm sure I'm fully qualified to do the same. It's not to say I don't question my own sanity for doing things the way I do at times, but the end-goal isn't employment; that's an entirely different issue, but to say the least, if I can't blow a MS-grad out of the water in the knowledge and utility sphere, I'll never have the confidence to do ask someone for money to do this stuff.

If I wanted to demonstrate my utility and "show I can ship" I'd push my other (possibly more) interesting things to my github, but what the hell, I don't really care at this point. I'm happy in my little learning and non-quitting bubble.

What's so different about integration compared to anything else?

Not sure if this is a serious question. Integration by parts, combined with trigonometric substitution and logarithmic identities should be enough to drive anyone up a wall... just don't forget to add a C at the end.

Plus, there is a certain class of integrals that don't have an answer at all. Imagine struggling to combine all the stuff, redoing and refactoring and trying from left to right and right to left, and that nice little "e) this integral doesn't have an answer" just staring back at you... and you know damn well some of the correct answers are e).

Has anyone taken Mathematical Biostats Boot Camp on Coursera? I need to brush up on Mathematical Statistics, and I'm looking for something somewhat more rigorous than most of the math-oriented classes I find on Coursera. Does this fit the bill?

I had to LOL at the FAQ:

Is calculus really necessary for this class?
Yes.

I would suspect this one would be pretty darn difficult then. I can't recall seeing a stat class that specifies a need to understand ODEs.

ha showoff.

but seriously, i don't think you're going to find classes as rigorous as your U of Chicago math classes. the point of those classes is to teach at a level that 50,000 random people can understand by putting in 10-15 hours per week of effort.


edit- I'm not knocking the classes there btw--i love them and hope they continue forever.

Some of these classes aren't watered down from the real class at all.

Calculus is required to make any sense of continuous random variables and their distributions, unless you're doing low-level stuff where other people did all the actual math for you in advance. I don't know if you've ever used a normal table, but that's basically just a bunch of precalculated variables for a very weird integral.

<== has a math/stats related degree

The software engineering classes are often pretty hard and interesting. For the theory classes, I don't think that the problem is that they want 50,000 random people to understand it, but that they can't grade 10s of thousands of problem sets saying "this implies that because..." or "this is a special case of that because ..." and actually analyze the arguments, and that's what forces them to sort of water it down.

so just started Stats 2.2x: Probability. Dives in much quicker than 2.1x (which was laughably easy to begin with). As a (former) poker player, a lot of this probability is already familiar to me, but they dive right into conditional probability, Bayes' Theorem, etc. Not particularly difficult, but if this is the first week it leaves me hopeful that this section will be more rigorous than the last.

Am also doing 2.2 but havent looked at week one yet. Glad to hear it might be a little tougher tho (little annoying that 40% is still a pass for this one). Excited for Machine Learning also...hoping I have enough time to dedicate to it.

I might sign up for the machine learning class as well. I'm in physics though and I was thinking about doing a coursera course that is somewhat different from what I do in my classes but it looks like a ton of matlab and stats so I may want a break from that over the summer. I think I will definitely sign up for the algorithms course that starts in June though.

What algorithms course that starts in June?

Crap never mind the algorithms class started in June 2012 . Coursera does list an algorithms class for 2013, but the dates haven't been announced yet. Hopefully it happens some time over the summer.

The Single Variable Calc class posted some statistics. Somehow expected and shocking at the same time:

For those of you that are curious, here are some numbers for the course:

Number of registered users: 47310
Number of submissions:
Diagnostic Exam: 8325
Chapter 1 Quiz: 2719
Chapter 2 Quiz: 1931
Chapter 3 Quiz: 1369
Chapter 4 Quiz: 981
Chapter 5 Quiz: 863
Final Exam: 835
Number of students that earned a certificate: 687

I had to stop at Chapter 4 quiz. I had an okay chance to get the cert but eh, I didn't bother with the exam for some reason, despite having two attempts. The next class starts on May 24. Definitely looking forward to doing this one again. The workload won't be so bad this time.

Id like to do it but dont think it's going to fit in...esp if it is really hard and esp cuz my toughest part of Calc One was applications which it sounds like this one is heavy on. Did see this on the Calc One forums from our prof.

But could be awhile still Id guess...here's hoping

Just starting to check this out. Looks like CS50x is finished (but still available) and 6.00x is still going but I already missed a bunch of deadlines. Is it still fairly valuable to do these courses on your own time? Somebody much earlier in the thread wrote that Malan is great, so I'd lean towards CS50x. Will it be taken down soon?

#newbquestions

The material is pretty much always available at cs50.net